Mutually Adversarial Self-Training with Evolving Data for Unified Multimodal Models
cs.CV, cs.AI
Submitted: 2026-09-28
Updated: 2026-09-28
Terminology
Sources
- Chameleon: Mixed-Modal Early-Fusion Foundation Models
- Next Token Prediction Towards Multimodal Intelligence: A Comprehensive Survey
- Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling
- Effective Reinforcement Learning for Agentic Search by Recycling Zero-Variance Queries During Training
- Emerging Properties in Unified Multimodal Pretraining
- SUDER: Self-Improving Unified Large Multimodal Models for Understanding and Generation with Dual Self-Rewards
- ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment
- Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback
- SRUM: Fine-Grained Self-Rewarding for Unified Multimodal Models
- SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension
- SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension
- DeepSeek-VL: Towards Real-World Vision-Language Understanding
- UniRL: Self-Improving Unified Multimodal Models via Supervised and Reinforcement Learning
- Dota 2 with Large Scale Deep Reinforcement Learning
- DUEL: Adversarial Self-Play for Multimodal Reasoning
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation
- A Survey on Self-Evolution of Large Language Models
- SimpleAR: Pushing the Frontier of Autoregressive Visual Generation through Pretraining, SFT, and RL
- Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models